A study on trend and modeling the export demand of canned pineapple in Malaysia by comparing between univariate modelling and box-jenkins methodology at Malaysian Pineapple Industry Board (MPIB) / Sunizan Jamhari

The purpose of this study is to determine a suitable method comparing between Univariate Modelling Technique and Box-Jenkins Methodology to forecast the export demand of canned pineapple from Malaysian Pineapple Industry Board (MPIB) using monthly data from January 2005 to October 2012. There are six models in Univariate Modelling Technique include Naïve with Trend, Single Exponential Smoothing, Double Exponential Smoothing, Holt’s Method, Adaptive Response Rate Exponential Smoothing Model (ARRES) and Holt’s Winter Model. While for the Box-Jenkins methodology, there are four models which are SARIMA(4,1,1)(1,1,1)12, SARIMA(3,1,0)(1,1,1)12, SARIMA(2,1,1)(1,1,1)12, SARIMA(1,1,1)(1,1,1)12. By comparing the value of Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE), Holt’s Winter give the best model for Univariate Modelling Technique while SARIMA(3,1,0)(1,1,1)12 give the best model for Box-Jenkins Methodology. By comparing the both best model, SARIMA(3,1,0)(1,1,1)12 give the best model. In addition it is shown that SARIMA(3,1,0)(1,1,1)12 fits the data well and they have correctly predicted the future trend of export canned pineapple within the sample period of study